Recently routing protocols for QoS aware multicast have been actively studied, but there are few studies focusing on the scalability of link state advertisement when the available bandwidth of a link is updated along with the QoS aware multicast tree construction. This paper proposes a new QoS aware multicast routing protocol that is scalable in terms of the link state advertisement exchange. Our protocol has the following features: (1) a multicast network is divided into domains, and the advertisement of information on links within a domain is limited within the domain; (2) among the border multicast routers, only the link state information of interdomain links is advertised, so that the number of link state advertisement messages is drastically reduced; (3) when a multicast tree spreads over multiple domains, the tree construction needs to be performed without information on links in other domains, and it is possible that the construction may fail. In order to cope with this problem, the crank back mechanism of tree construction is introduced. This paper describes the detailed procedures and the message formats of our protocol. It also describes the evaluation of the number of exchanged link state advertisement messages and shows that our protocol can reduce the number by the order of 1/(number of domains) compared with conventional protocols.

Recently audio and video delivery services are widely spread in the Internet. In order to deliver these data to multiple receivers at the same time, the multicast technologies are indispensable. In such a situation, the performance analysis of multicast traffic will be important in order to design and operate the Internet. Multicast traffic is controlled by its own routing protocol and the multicast routing protocol is more complex than conventional unicast routing protocols. In this paper we propose a new multicast performance analyzer considering distance vector multicast routing protocol (DVMRP) and protocol independent multicast-sparse mode (PIM-SM) routing protocols. It captures messages of a multicast routing protocol as well as multicast IP datagrams. It also analyzes DVMRP and PIM-SM messages in order to investigate the behaviors of multicast IP datagrams and measures the statistics of multicast traffic in detail. This paper describes the detailed design of the multicast performance analyzer and the results of analyzing multicast datagrams transmitted over networks.

Most of current video retrieval systems use video transfer protocols such that servers simply transmit video packets at the same rate as clients play them. If any packets are corrupted during transmission, they will be lost and cannot be recovered by retransmission. In video retrieval systems however the video data are stored in servers and clients can prefetch them prior to playing. So, it might be possible for the video retrieval systems to make corrupted video packets retransmitted before the play-out deadline. However, the application of existing reliable protocols causes problems such that, if a packet does not arrives before the deadline due to retransmission, the packets following it will not be delivered to the upper layer even if they have already arrived. We propose a new video transfer protocol for video retrieval systems over an ATM network which provides video data prefetch, flow control for the video buffer, selective retransmission with a skipping function for video packets that are late for the play-out deadline, and a resynchronization function for the video buffer. We have implemented an experimental system using our protocol and evaluated the performance. The results of the performance evaluation shows that the proposed protocol decreases the number of unplayed video data to less than 1/70 compared with the conventional non-retransmission protocol when random bit errors with a BER of 10/sup -6/ are inserted in an ATM network.

Along with the rapid progress of optical technologies, the transmission speed of LANs and public networks has been increased significantly. As a result, it has become possible for computers distributed geographically to communicate with each other with high throughput. However, the current protocols such as TCP/IP have some problems, especially the performance through long distance and wide bandwidth networks. Therefore, new protocols with new data transfer algorithms are required. In this paper, we describe an implementation of a high speed transport protocol by a user level library which interfaces to UDP/IP. This method uses only functions commonly provided by UNIX operating systems, and therefore, allows a new protocol to be developed easily and to be ported to other UNIX workstations easily. Our library called the transport library realizes the coordination of buffer managements in the application and the library, and low overhead and prompt handling of receive and timer interrupts in order to achieve high performance. Our implementation results show 32 Mbits/sec over an ATM network whose effective transmission speed is 36 Mbits/sec, regardless of the propagation delay from 0 to 200 msec. These values of throughput are better than those of the in-kernel TCP programs.